sidebanner.jpg (5330 bytes)

wpe27.jpg (5472 bytes)

wpe28.jpg (6821 bytes)

wpe29.jpg (6631 bytes)

A Quick and Easy Guide to RAID Technology

Have you ever passed up the chance to quote on a server simply because it asked for a RAID solution and you were unsure of how to implement it? EDsys Computers are here to help you implement effective RAID solutions to suit your individual needs.

Who should consider using RAID? Any business that requires a high level of data integrity and continuity of service from their server should seriously consider a RAID solution. RAID is also suitable in situations where performance is critical. RAID technology is no longer an expensive option, and should be considered by any business that values data security and performance.

So, what is RAID? RAID is yet another acronym in the industry and this one stands for ‘Redundant Array of Inexpensive Disks’. While some people may argue that a 9Gb SCSI drive is hardly ‘inexpensive’, that word is often accurately substituted with ‘independent’. Either way, it means the same.

RAID Technology was first defined by a group of computer scientists at the University of California at Berkeley in the late 80’s. The idea behind RAID started with the possibility of using two or more disks to appear as a single device to the host system in order to enhance performance over a large, single-disk storage device.

 

Performance was better, but reliability was effectively halved. So the need for some form of redundancy was created. And thus they went from JBOD (Just a Bunch Of Drives) to RAID. While there are several different levels of RAID, the most popular are RAID 1 and RAID 5.

RAID 1 is a popular small scale scenario whereby all data is duplicated on two drives. This is very effective in redundancy, yet expensive for storage as extra drives need to be installed two at a time. This method is also known as ‘mirroring’.

wpe1A.jpg (5798 bytes)
 

In a RAID 5 situation, redundancy works by Block-level data striping with distributed parity. Essentially, parity is the addition of all the drives used in the array. Recovery from a drive failure is achieved by reading the remaining good data and checking it against the parity data stored by the array. In other words, using a minimum of three drives, the total storage capacity is that of two drives, although the parity is distributed across all three. So by adding another drive for example, your storage percentage goes from 66% to 75%. By adding another drive again we go to 80% and so on. Think of it as a car with a spare tyre, you only need one spare, regardless of how many tyres you actually have on the car. This solution is relatively inexpensive to expand as extra drives can be added singularly, providing they have the same storage capacity.

RAID Level 5:

wpe39.jpg (8163 bytes) wpe3A.jpg (6293 bytes)
 
To download this document in Acrobat format. Click Here (112KB)